Ubiquitin is a small, monomeric, and cytosolic protein which is highly conserved in sequence and is present in all known eukaryotic cells from protozoans to vertebrates. In the organism, it plays a crucial role in the regulation of the controlled degradation of cellular proteins. For this purpose, the proteins destined for degradation are covalently linked to ubiquitin or polyubiquitin chains during their passage through a cascade of enzymes and are selectively degraded because of this label. According to recent results, ubiquitin or the labelling of proteins by ubiquitin, respectively, plays an important role also in other cellular processes such as the import of several proteins or the gene regulation thereof (Marx, 2002).
Besides the clarification of its physiological function, ubiquitin is a research object primarily because of its structural and proteinchemical properties. The polypeptide chain of ubiquitin consists of 76 amino acids folded in an extraordinarily compact alpha/beta structure (Vijay-Kumar, 1987): almost 87% of the polypeptide chain are involved in the formation of the secondary structural elements by means of hydrogen bonds. As prominent secondary structures can be mentioned three and a half alpha-helical turns as well as an antiparallel beta-sheet consisting of four strands. The characteristic arrangement of these elements—an antiparallel beta sheet exposed to the protein surface onto the back side of which an alpha helix is packed which lies vertically on top of it—is generally considered as so-called ubiquitin-like folding motif. Therefore, ubiquitin is name-giving for the respective protein superfamily (“ubiquitin-like proteins”) or the protein family (“ubiquitin-related proteins”), respectively, (Murzin et al., 1995) which comprises proteins such as for example SUMO-1 (Müller et al., 2001), FAU (Michiels et al., 1993), NEDD-8 (Kumar et al., 1993), UBL-1 (Jones and Candino, 1993), and GDX (Filippi et al., 1990) bearing this motif as well as a high degree of identity to ubiquitin in their primary sequence.
Human Ubiquitin is—as mentioned above—a 76 amino acid long polypeptide (FIG. 1). It is a small, globular protein of 7.5 kDa with an extruding C-terminus. The main structural features of ubiquitin are shown in FIG. 1. Hydrophobic residues from the sheet and from the helix assemble the hydrophobic core of the protein and stabilize the orientation of the helix (FIG. 2). The highly dense hydrophobic packing of the protein's core is reflected by its excellent thermodynamic stability, which should make the protein to an ideal candidate to use it as a scaffold for protein engineering approaches.
Because of its small size, the artificial preparation of ubiquitin can be carried out both by chemical synthesis and by means of biotechnological methods. Due to the favourable folding properties, ubiquitin can be produced by genetic engineering using microorganisms such as Escherichia coli in relatively large amounts either in the cytosol or in the periplasmic space. Because of the oxidizing conditions predominating in the periplasm the latter strategy generally is reserved for the production of secretory proteins. Due to the simple and efficient bacterial preparation ubiquitin can be used as a fusion partner for other foreign proteins to be prepared for which the production is problematic. By means of the fusion to ubiquitin an improved solubility and thereby an improved yield can be achieved. The approach practised in the present invention to provide ubiquitin as universal artificial binding protein allows for a completely novel utilization of its proteinchemical properties.
Among those proteins of which the natural function is utilized in diagnostics and pharmaceutics the immunoglobulins play a predominant role. Their ability of specific, non-covalent binding to a wide range of different substances makes them the most important tool for bioscientific application. The methods developed in recent years for the functional biosynthesis of antibody fragments in E. coli have further extended the possibilities using immunoglobulins but have at the same time demonstrated their limitations.
Besides Fab- and Fv-fragments (Skerra and Plückthun, 1988) which principally can be obtained by conventional methods, different artificial constructs could be developed by means of protein engineering methods. Assisted by the modular structure of immunoglobulins (reviewed in Düthel and Kontermann, 2001), notably single chain Fv fragments (scFv) (Bird et al., 1988), disulfide-bridged Fv fragments (dsFv) (Brinkmann et al., 1993) as well as bivalent (Carter et al., 1992) and bispecific antibody fragments (e.g. diabodies, Holliger et al., 1993) could be generated. For diagnosis and the use in therapy bifunctional proteins can be obtained by genetic fusion of the recombinant Ig fragments to effector modules. Thus, fusions to alkaline phosphatase (Müller et al., 1999) and the green fluorescent protein (GFP; Griep et al., 1999) are available among others. Fusions of antibody fragments to radioisotopes or cytotoxic substances are of great potential importance for cancer treatment (immunotoxins; Reiter and Pastan, 1998). In this case, the selective binding of respective Ig fragments to specific surface proteins on tumor cells is utilized for the site-specific application of therapeutics (tumor targeting).
However, the methods for the preparation of antibody fragments in E. coli not only allow for their provision for diagnostics and therapy in sufficient quality and quantity but also for simple and quick modification of their protein- and immunochemical properties. The easy handling of a bacterial host enables a straightforward alteration of the vector-encoded genes for the foreign protein by means of standard molecular-biological methods. By means of a targeted antibody engineering (Kontermann and Dübel, 2001) antibody fragments can thus be optimised e.g. with respect to their binding affinity or their host compatibility. Also, specific antibodies or fragments thereof, respectively, can be prepared artificially, i.e. out of the immune system, which are directed against the most different target substances such as low molecular weight structures or proteins for example. By such evolutive methods synthetic libraries of antibody fragments are prepared by the introduction of random mutations which in their extent can be close to the human repertoire (Knappik et al., 2000). By means of suitable selection strategies such as phage display or ribosome display (Winter, 1998, Hoogenboom et al., 1998; Hanes et al., 2000) functional Ig fragments having the desired binding property are isolated in the case of success. In this manner it is also possible for example to obtain binding proteins for such antigens which during a classical immunization would provoke toxic effects or only a weak immune response.
Despite the above-mentioned achievements and possibilities provided by antibody engineering certain disadvantages can limit the practical use of antibodies. Thus, it is a problem to provide them in sufficient amounts.
The production of functional antibodies is carried out primarily in eukaryotic cell culture systems, which is an extraordinarily cost-intensive method. Furthermore, the low tissue penetration of the antibody molecules due to their size and their long residence time in the serum (slow blood clearance), respectively, hamper many therapeutic applications. Although smaller fragments of antibodies such as scFv or Fab fragments (see above) can be prepared in bacteria and thus basically at lower costs the yields of this recombinant production, however, are lower than the desired level due to their unfavourable folding properties and the required formation of several disulfide bonds. Moreover, recombinant antibody fragments often show a reduced thermodynamic stability, a lower binding activity and higher aggregation tendencies, as compared to the parental antibodies. In order to circumvent such limitations attempts have been made to impart the principle of antibody binding—namely the binding by means of a hypervariable surface-exposed region localized on a conserved protein scaffold—to other proteins (Skerra, 2000). This means that essentially variable loops are varied in order to generate an artificial binding property. For this purpose, usually natural binding proteins such as e.g. lipocalins (Beste et al., 1999) or the fibronectin type III domain (Koide et al., 1998) have been used as a starting point for which binding sites are formed in a manner analogously to antibodies from flexible “loop” structures whose modification enables the induced fit recognition of ligands.
WO2004/106368 relates to modified proteins of the superfamily of “ubiquitin-like proteins”, proteins that have a ubiquitin-like fold. As a result of said modification, the proteins have a binding affinity with respect to a predetermined binding partner that did not exist previously. The invention also relates to a method for the production and utilization of said proteins. According to WO2004/106368 a protein is provided selected from the group consisting of proteins of the protein superfamily of “ubiquitin-like proteins” wherein due to one or more modifications of amino acids in the beta sheet strand of the beta sheet region the protein shows a modified binding affinity. Thus, a de novo binding site on human ubiquitin was assembled out of new amino acid substitutions at the positions Q2, F4, K6, Q62, K63, E64, S65 and T66 (FIG. 6, 7).
Unfortunately, due to the close proximity of the de novo binding site to the N-terminus of the protein, the codon composition of the newly introduced amino acids resulted in inhomogeneous expression rates of the protein variants and severely constricted a subsequent N-terminal genetic fusion or posttranslational N-terminal labeling of the variants. Finally, the development of high affinity binding polypeptides using the convex beta sheet topology as a paratope was dependent on the presence of a concave structure of the target molecule.